ClaimsGary Marcus

“I don’t expect my predictions to be as on target this year as last, for a happy reason:”

Gary Marcus, author January 2026 Six (or seven) predictions for AI 2026 from a Generative AI realist (Marcus on AI)

Prediction Their own work or organisation

Gary Marcus does not expect his 2026 predictions to be as on target as his 2025 predictions were.

Not yet due The date it refers to has not come.

Prediction: Statements about events outside the person's control, checked once they fell due: did they come true?

Reasoning

deadline 2026 is after 2026-11-04

Cite this claim

Gary Marcus, “Gary Marcus does not expect his 2026 predictions to be as on target as his 2025 predictions were.” (said January 2026). Verdict: not yet due. Reliability API, claim c_2c5390abc985, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_2c5390abc985

This claim is in dataset release v1.1.2, with the same verdict.

Reply to this claim

For Gary Marcus, or someone speaking for them: a correction or counter-evidence. Replies are read before anything is shown; an approved reply shows your name, role, date, statement and link. Your email address is never shown and is deleted 30 days after the decision (privacy).

More prediction claims by Gary Marcus

  1. Mostly true Prediction 27 July 2025
    “Some of its advice will be head-scratchingly bad.”

    Some of GPT-5's advice will be very bad.

  2. Mostly true Prediction 27 July 2025
    “GPT-5 will surely be better, a lot better than GPT-4.”

    GPT-5 will be considerably better than GPT-4.

  3. Resolved true Prediction 27 July 2025
    “GPT-5 really is about to drop, no foolin’.”

    OpenAI will release GPT-5 imminently.

  4. Resolved true Prediction 27 July 2025
    “It will still make a significant number of shake-your-head stupid errors, in ways that are hard to fully predict.”

    GPT-5 will still make a significant number of obvious errors that are hard to predict.

  5. Resolved true Prediction 27 July 2025
    “Guardrails (a la ChatGPT) may be in place, but the guardrails will teeter between being too weak (beaten by “jailbreaks”) and too strong (rejecting some perfectly reasonable requests).”

    If the model being discussed has guardrails like ChatGPT's, they will teeter between being too weak (beaten by jailbreaks) and too strong (rejecting some perfectly reasonable requests).

All of them

Verdicts are made automatically; how verdicts are made.